The host-feeding patterns of Anopheles albimanus Wiedemann were described and the effect of host availability on these patterns was assessed in three different ecological areas of coastal Chiapas, Mexico. Resting mosquitoes were collected indoors and outdoors during rainy seasons. A 20% sample of blood-fed mosquitoes was tested to determine the source of the blood meal using an ELISA. The unweighted human blood index (HBI) of An. albimanus in the three areas ranged from 0.11 to 0.21. This mosquito species fed more frequently on bovines than on any other host, but the forage ratio indicated that there was also a high preference for equines. Some females tended to rest or complete their gonotrophic cycle indoors after feeding on animals, but females also fed on man and rested outdoors. Host availability and ecological conditions appeared to be responsible for differences observed in the HBI among areas.

No abstract

Responses of the malaria vector Anopheles albimanus to pyrethroid impregnated bednets made of cotton or nylon, compared with untreated nets, were investigated in houses occupied by two people inside and/or outside two bednets, in coastal Chiapas, México. The pyrethroid used was lambdacyhalothrin 30 mg a.i./m2. Bioassay mortality rates of An.albimanus exposed to treated nets for 3 or 15 min, rose from 40-55% to 90-100% for nylon nets 3-19 weeks post-treatment, but were consistently lower for treated cotton nets. An.albimanus females (collected unfed on human bait) were released in houses surrounded by curtains for trapping mosquitoes that exited from the house. Floor sheets were used in and around each experimental house for retrieving any mosquitoes knocked-down and/or killed. During post-treatment assessment for 17 weeks, An.albimanus blood-feeding success rates were 23-24% with untreated nets, 14-18% with treated cotton nets and 8-15% with treated nylon nets, significantly reduced when both human baits were inside the treated bednets, but not when one or more baits were outside the treated bednet(s) within the house. Proportions of mosquitoes leaving houses < 3 h post-release were 53-59% from houses with untreated bednets versus 65-78% with treated bednets. Except in one case (when both humans were outside treated cotton nets), these increased early exit rates were significant, whether or not the human baits were inside the treated bednets indoors. Mortality rates of An.albimanus females exiting overnight (22.00-06.00 hours) averaged 15-39% from houses with treated cotton and 16-46% with treated nylon nets, very significantly greater than the control mean rates of 6-8% mortality with untreated nets. Observations on wild-caught An.albimanus females marked with fluorescent powder and released indoors revealed that few mosquitoes (3-11%) actually contacted the bednets unless both human baits remained under them - when contact rates were 22% on treated nylon, 23% on treated cotton and 42% on untreated nets (P approximately 0.05). The mean resting time was significantly longer on untreated (14.4 min) than on treated nylon (5.8 min) or cotton (9.5 min) bednets, whereas mean resting times on other surfaces indoors were 16.5-19.8 min. Proportions exiting within 2h of release were significantly more from houses with treated houses (33-35%) than with untreated nets (8%). However, mortality rates of mosquitoes that landed on treated nets were very significantly greater (90-100%) than after landing on untreated nets (10%). Thus, despite some excito-repellency, lambdacyhalothrin-impregnated bednets (especially made of nylon) proved to be effective as an alternative to house-spraying against An.albimanus.

The existence of sympatric populations within Anopheles albimanus Wiedemann with specific host preferences and the question of whether host selection has a genetic basis or is a learned characteristic were investigated. Progeny of wild females collected from corrals or human bait were reared in an insectary. F1 females were fed on a cow or on a human host, allowed to oviposit, and then were released in an experimental hut divided into three compartments. The side rooms occupied by two human or one calf baits were connected by a "funnel-like" cone with the central room in which mosquitoes were released. Females were released at 2000 hours and recaptured the next morning at 0600 hours. A fixed proportion of mosquitoes selected either host (approximately 65% cow and 35% human), irrespective of their parental origin or source of the first blood meal. The data indicate that the existence of cryptic populations with anthropophilic habits is unlikely and that a "learned" host selection also may be ruled out.

The length of the gonotrophic cycle of Anopheles albimanus was estimated by 12 mark-release-recapture studies conducted in corrals in southern Mexico from 1987 to 1990. The initial set of three mark-release studies indicated that the gonotrophic cycle takes at least 4 d based on the day when gravid mosquitoes were first recaptured. However, in later experiments, mosquitoes recaptured seeking hosts at 48 h after release were in Sella's and Christophers' stages I and II, but parity rates had nearly doubled, indicating that eggs may have developed in less than 48 h and that mosquitoes returned to refeed immediately following oviposition. Two gonotrophic cycles probably exist, one of 48 h dominated by parous mosquitoes and one of 4 d comprised of nulliparous, pregravid mosquitoes. Daily survivorship was estimated by regression from the decrease in the daily recapture rate (0.46-0.68) was less than that estimated by the parity rates (0.67-0.69), with no important differences found between wet and dry seasons. Estimates of the probability of a mosquito living long enough to transmit malaria were 2% (range, 1.8-2.5%).